1. Field of the Invention
The present invention relates to electric motors and, more particularly, to direct current electric motors suitable for reliably and efficiently powering electric vehicles and industrial machinery.
2. Description of Related Art
There are numerous electric motors available for propelling electric automobiles. These include both direct current (DC) motors designed to drive directly off of the batteries and alternating current (AC) motors which require electrical circuitry for converting the DC power in the batteries to AC power. The most efficient of these AC motors requires three or more phase power.
Such motors have a high power-to-weight ratio, can be made to run efficiently, and are inherently reliable because of their brushless design. A disadvantage of such motors is the fact that the battery power must be first converted to AC before it can be used by the motor. This disadvantage shows up in the need for complex circuitry. This is especially true for AC motors having three or more phases. Along with the need for complex circuitry is the fact that the failure of even a single electrical component in the system can result in total failure of the drive circuitry for producing AC power. This results in DC electric power in the batteries, and a motor that requires AC power. This renders the entire drive system useless.
Therefore, such drive systems suffer from the potential of leaving the driver stranded. Despite these obstacles, companies such as AC Propulsion Inc. have made considerable advances in the use of AC motors in electric cars. In particular, high power to weight ratios have been achieved.
In the powering of industrial machinery, in many applications it is desirable to have an electric motor that has considerable amounts of torque and power at relatively low RPM values. This is normally achieved by gearing the motor down, however this practice results in added moving parts, increased mechanical losses, and adds cost and complexity to the overall system. In general DC electric motors have good torque characteristics which make them ideal for use in many industrial applications. In general with DC electric motors, the more mechanical drag on the motor, the more torque is produced. In this respect such motors are ideal for propelling electric cars as well. This is especially true if one wants to drive the wheels of such a vehicle directly by employing a motor in the wheel hub. There are several reasons why DC electric motors are advantageous.
DC electric motors require little circuitry to drive them from batteries. In some cases they can be wired directly with only a switch to turn the system on and off. Another advantage offered by DC electric motors is the fact that such motors do not require starting circuitry in the way that many AC motors do.
The first electric cars were produced at the turn of the century and were powered by DC electric motors. Such motors utilized two sets of electromagnets to produce their torque. One set was mounted to the inside of the motor casing. These electromagnets had one set of poles facing inward, and the other set of poles against the steel casing to magnetically connect them in series. The motor casing with its electromagnets made up the stator portion of the motor. When power was on, these electromagnets maintained the same field. At each end of the motor casing were end caps having holes which were centrally located which supported a bushing or bearing through which the rotational portion (or rotor) was supported. The rotor consisted of a round shaft having a larger diameter set of electromagnet windings wound onto an iron core. The ends of the rotor windings were fixed to conductive copper strips that were insulated from each other and the motor casing using resin or other suitable insulating material. A set of brushes which were usually made of graphite pushed up against the copper strips in the rotor to make electrical contact while also allowing the rotor to rotate. The position of the brushes relative to the stator electromagnet windings was always set so that the proper rotor electromagnets were turned on at the appropriate times by the brushes to always magnetically drive the rotor in the same direction (i.e., by interaction of the stator magnetic field with the magnetic field of the electromagnets in the rotor).
While these motors were suitable for powering both electric cars, as well as industrial equipment, Their efficiency was somewhat limited by the fact that power losses occurred in both sets of electromagnets due to the resistance of their windings.
In the early 1930s, the General Electric Company developed the first permanent magnets that were strong enough to replace one set of electromagnet windings in DC motors. This Permanent magnet material was called Alnico, and soon several grades were made commercially available. Shortly thereafter, the first useful permanent magnet motors began to appear. These motors basically used permanent magnets to replace the stator electromagnets. While these motors had an increased efficiency when compared to their predecessors, they suffered from the possibility of demagnetization of the permanent magnets if the electromagnetic field in the rotor exceeded the xe2x80x9ccoercive forcexe2x80x9d (a measure of the resistance to demagnetization of permanent magnets) of the permanent magnets in the stator. To partially alleviate this problem, stronger ceramic permanent magnets were developed, and still stronger magnets called xe2x80x9cRare Earth Magnetsxe2x80x9d are among the most recent developments.
All DC permanent magnet motors run the risk of demagnetization of their permanent magnets if the electromagnetic field of the windings exceed the coercive force of the permanent magnets. To alleviate this problem, a maximum safe operating voltage for any DC permanent magnet motor is specified which under maximum power conditions (i.e., at stall) the resistance of the electromagnet windings will be high enough to prevent a flow of current through the electromagnet sufficient to cause irreversible damage to the permanent magnets. This current is considerably greater than the normal operating current, and for this reason, normal operating conditions for traditional permanent magnet DC electric motors only utilize a fraction of their true power capabilities based on their permanent magnets. In fact, most of these motors only utilize between 10% and 25% of their true potential.
The wire diameter used in winding an electromagnet core basically determines the magnetization force in ampere-turns for a given cross-sectional core diameter at a given voltage. Increasing the number of turns reduces the number of amperes that will flow through the coil, but increases the number of turns thus, maintaining the same number of ampere-turns. In order to more effectively use the permanent magnets of a permanent magnet motor under normal running conditions (i.e., at 10% to 25% of stall current) electromagnet windings must be activated that are more than capable of demagnetizing the permanent magnets in the motor under the conditions of stall. One way to accomplish this is to wind the electromagnet in layers and using thinner wire successively in the outer layers. On start up, all the layers of wire in the electromagnet are used. The resistance of the thinner outer wire prevents excessive currents in the motor thus preventing demagnetization of the motor permanent magnets. Once the motor RPM value reaches a safe level, the outer layers are shunted, thus increasing ampere turns in the motor and increasing the utilization of the motor permanent magnets. An interlock is also provided that prevents accidental activation of the shunt mechanism under stall or low RPM conditions. Other methods may be employed to wind electromagnet cores providing this type of electromagnet. Also included is twisting two or more strands of a given thickness insulated electromagnet wire together and winding the core. At one end the two strands are connected. This becomes the common. At the other end, the two leads are kept separate. On motor start up, only one lead is connected. Once motor RPM values reach a safe level, the second lead is connected to the first thereby increasing the utilization of the permanent magnets by providing more ampere-turns in the motor electromagnet assembly.
This increased utilization of permanent magnets during running conditions results in an electric motor having exceptionally high torque values. The example outlined below will be used to further illustrate this point.
Ametek Corporation is a major manufacturer of electric motors. They manufacture a DC permanent magnet motor which is nominally rated at 24 volts. #116281-00. It is reversible and has a no load speed of 980 RPM. Measurement of the armature resistance under stall conditions gives 2.4 ohms. According to the manufacturer""s specifications, it is desirable to run this motor at 750 RPM. This value represents about 25% of stall torque at 24 volts input. Under these conditions, 2.75 amperes of current flow through the motor windings. The voltage drop is equal to the Resistance in ohms times the current in amperes. This is equal to 2.4 ohmsxc3x972.75 amperes=6.6 volts. This leaves 17.4 volts for contributing to mechanical work giving the motor an efficiency of about 73%. The mechanical output when using this motor under the manufacturers recommended voltage, torque, and RPM values is 2.75 amperesxc3x9724 voltsxc3x970.73 efficiency factor to give 48 watts or about 0.063 horsepower. Running under the conditions of less torque increases motor efficiency however is detrimental to power output. Loading down the motor to increase torque has the effect of trading torque for RPM which results in virtually no increase in mechanical work output accompanied by a rapid increase in power losses in the motor windings.
Accordingly, The manufacturer also provided the maximum current before demagnetization occurs in the motor. This occurs at a current of 20 amperes through the windings. Note the fact that up to 20 amperes can be delivered to the motor before damaging the permanent magnets, but the manufacturer also specifies the best load value to run the motor at is 2.75 amperes. Therefore during normal running conditions, only 13.7% of the capability of the permanent magnets are utilized. This indicates that under the proper conditions that this motor can put out 7.3 times this torque value at this RPM value before demagnetizing the stator permanent magnets.
As is, this motor can withstand 20 amperes through its electromagnet windings before demagnetization effects destroy the stator permanent magnets. With an armature resistance of 2.4 ohms (measured at stall) this gives a maximum safe operating voltage of 48 volts without the risk of damaging the stator permanent magnets. Of course other factors need to be considered before exceeding the manufacturers specifications, but this electric motor can handle stall at 48 volts before demagnetizing the stator permanent magnets. This gives a power rating considerably higher than the original manufacturers specifications. If one wants to run this particular motor at 48 volts and at 25% of stall torque (Maximum mechanical work output before excessive voltage drops across the windings occur) at 48 volts, and about 5 amperes, 240 watts of power will be delivered to the motor at 1,500 RPM with an approximate efficiency of 75%, producing 180 watts of mechanical work and 60 watts of waste heat being generated in the motor windings. In this case the permanent magnet utilization during running is 25% of their true capability. If one wants to increase the torque output at this RPM value, it is necessary to further utilize the stator permanent magnets during running. This can be accomplished using motor windings in the armature that are more than capable of demagnetizing the stator permanent magnets under stall conditions, and not fully activating these windings until the motor RPM reaches a safe value that will not demagnetize the stator permanent magnets. The following theoretical analysis will illustrate this point.
An arbitrary number of 100 turns is chosen for the wire that wraps the electromagnet core. This gives a value of 2,000 ampere-turns before demagnetization occurs. A wire diameter is chosen to wrap the electromagnet rotor core that is half the resistance of the original wire. If the same number of turns are used, (100) an armature resistance of 1.2 ohms would be the result. Since the same number of turns is used, if 48 volts is applied to the armature windings under stall conditions, 4,000 ampere-turns would result. Twice the magnetic field needed to demagnetize the stator permanent magnets (assuming that core saturation effects are negligible). Now this particular motor as is would not be suitable for running at 48 volts because if even 50% of stall torque is achieved, the stator permanent magnets will be at least partially demagnetized. A second set of electromagnet windings is employed in series with the first set of windings using wire of three times the electrical resistance as the first original set. This set of windings is then wound with the same number of turns (100) as the first set giving a resistance of 3.6 ohms. The total resistance of the two windings combined is 4.8 ohms. Since the electromagnet core has twice the number of original turns, (200) in terms of demagnetization (2000 ampere-turns) this motor would have a safe operating current of 10 amperes. With an armature resistance of 4.8 ohms, a maximum operating voltage of 48 volts can be employed under the conditions of stall without demagnetizing the stator permanent magnets.
Now assume that at start up, this particular motor is connected to a 48-volt power source. At start up, all of the windings are activated causing a current of 10 amperes to flow. 2,000 ampere-turns results in the electromagnet which is just shy of the magnetic field needed to demagnetize the stator permanent magnets. The rotor under this strong magnetic field starts to spin rapidly, and the current along with its associated magnetic field starts to weaken. At some RPM value, it is now safe to shunt out the outer high resistance turns of wire thus activating only the lower resistance inner turns. Since 25% of stall current is the desired operating range, and the rotor electromagnets have the same number of turns as the original motor, at 1,500 RPM the motor efficiency is 75%, but the torque is twice the value as the original motor, and the permanent magnets are being utilized at 50% instead of 25%. This motor is now putting out twice the power as it was originally. 480 watts of input power is providing 360 watts of mechanical work accompanied by 120 watts of waste heat. This method of increasing the power output of permanent magnet motors is quite effective. This process however needs to be made user friendly. This requires that the end user of such motors does not have to think about damaging such motors by fully activating the lower resistance motor windings. In order to accomplish this end, interlocking circuitry is needed in order to prevent the accidental demagnetization of permanent magnets in these permanent magnet motors.
Another example of how this approach can yield positive results would be to employ the thick inner windings of the previous example with its thinner associated outer windings, but only run the motor at 12.5% of stall instead of 25% of stall. In this way, this electric motor would utilize the permanent magnets by only 25%, the same utilization as the original motor without these modified windings. The motor efficiency, however would be significantly improved from about 75% to 88%. As now the thicker electromagnet windings would have less resistive losses. Of course the same precautions to guard against demagnetizing the permanent magnets would have to be taken.
In these particular cases, the connections to the electromagnet windings are difficult to achieve because the electromagnet is embedded into the rotary part of the motor. Not that this cannot be achieved; it can, in at least two different ways.
In the first method, two different and complete sets of both brushes and commutators may be employed. This is not a preferred method of implementing the high magnetic utilization aspects of this invention. Brushes and commutators are a major source of motor wear and maintenance problems. Adding on another set would add complication and increase ,the overall need for motor maintenance. A better approach would be to employ an internal centrifugal switch that switched the motor windings only when the RPM value reached a safe level. Such switches are commonly used to lock out motor starter windings in many single phase AC induction motors. Although two methods of using this system are described above, it is far easier implemented in DC permanent magnet motors of brushless design. In such electric motors, the electromagnets make up the stationary part of the motor or stator, and the permanent magnets make up the rotating portion of the motor or rotor.
Brushless DC electric motors are permanent magnet motors which employ their permanent magnets into the rotating portion (rotor) of the motor and their electromagnets into the stationary portion (stator) of the motor. This results in a design which is inverted from traditional permanent magnet motors. Because the electromagnets used in such motors are stationary, no electrical power needs to be provided to any moving parts. This elimination of brush timed commutators is accomplished by switching arrangements which are comprised of a rotor position sensor and transistor switching circuitry. Several methods may be employed for sensing rotor position including Hall-effect sensors, and photocell gates. Transistor switching circuitry usually consists of transistors for the actual switching of power to the stationary electromagnets along with diode protecting circuitry which protects the transistors from transient back voltage spikes that are often associated with the rapid switching of electromagnets.
Another alternative is to construct a DC permanent magnet motor having the permanent magnets in the rotor and the electromagnets in the stator as is done with brushless designs but use a brush system for timing. The timing signal is then amplified electronically using transistors. This eliminates many of the problems in brush timed motors by significantly reducing the power that flows through the brush commutator system. This allows the brushes to be made from such materials as conductive kapton polyimide film, and also for the use of commutators made from copper clad circuit board material. With this type of hybrid brush timing system, keeping the inverted design features normally found in brushless motors is beneficial when employing multiple winding electromagnets.
Brushless permanent magnet motors are ideal for increasing permanent magnet utilization during running by employing the teachings of this invention. With the electromagnets in the stator instead of the rotor, the use of multiple electromagnet windings is easily achieved. In addition since the switching of the electromagnets is now achieved using electrical switching circuitry, interlocking of the individual electromagnet windings becomes a relatively easy task to perform. Speed sensing is critical as it is now used as a determining factor in the interlocking of various electromagnet-winding configurations. This serves the all-important function of protecting the permanent magnets from accidental demagnetization. Several speed sensing and interlocking approaches may be used. For example, for relatively large motors, a small electric generator may be placed onto the power output shaft of the motor and wired to a relay which closes the appropriate contacts at a predetermined generator output voltage. This system is a good choice because if any of the speed sensing components fail, the high power electromagnet windings are interlocked out and the motor while not being able to go into the high power mode will not be damaged by the accidental premature activation of these high power electromagnet windings. Another similar option for speed sensing is to place a small coil of wire on a ferromagnetic core which is placed close enough to the permanent magnet rotor to generate an output voltage which is proportional to speed. This sensor then becomes the electromagnet portion of a permanent magnet generator which uses the rotating rotor permanent magnets as its own.
According to another aspect of this invention a high power DC electric motor which is suitable for powering electric automobiles as well as industrial machinery is provided.
This particular electric motor describes the use of a large diameter planar rotor employing built in air moving vanes to provide cooling to stator electromagnets. All electric motors have resistive losses in their electromagnet windings which generate considerable amounts of heat. Because of this generation of heat, many motors both AC and DC, are equipped with a small fan blade mounted onto the motor shaft on the inside of the motor to move air through the motor for the purposes of cooling. In the case of DC motors, the greatest amount of heat is generated under high load conditions. Because of this fact, coupled with the fact that under such conditions motor RPM values are low, insufficient volumes of air are available to adequately cool the motor. This leads to the undesirable risk of overheating the motor under heavy load conditions. Because of this, it is often practice to mount an external fan powered by a separate power source to continuously blow high volumes of air through the motor. Although efficient for motor cooling under low RPM heavy load conditions, the extra fan motor adds to the complication of the system.
In addition to the overheating issue, the overall power of DC permanent magnet motors is proportional to the amount of permanent magnet material that can be magnetically cycled through the field produced by the electromagnets. To achieve a high rate of magnetic cycling at relatively low RPM values, employing a large diameter rotor is beneficial.
Increasing the rotor diameter increases the surface speed at the edge allowing a high volume of airflow to be easily achieved by adding air-moving surfaces about the periphery. This airflow is needed to cool the motor. For high power output applications, further cooling may be necessary. Blades, rotary vanes, or even an internal turbofan can be added to achieve this.
Several electromagnet to permanent magnet geometries may be employed. For example, the permanent magnets may be placed facing outward in a radial configuration at the periphery of a large diameter disc. With this permanent magnet configuration, the electromagnets are placed around the periphery with their poles facing inward in a radial direction. In this particular situation, rotary vanes could be added around the periphery on the top side and/or the bottom side of the permanent magnets for the purposes of moving air over the exposed top and bottom surfaces of the electromagnets in the stator.
Another possibility is to use a star shaped electromagnet in the center, and use a ring of permanent magnets that travels on the outside periphery of the centrally located electromagnets. In this configuration it is best practice to employ a central turbofan design to the rotary portion which pumps air over the central windings from top to bottom. Another geometry which can be employed is to place the permanent magnets into the periphery portion of a large diameter flat disc with their direction of magnetization transversing through the disc. The desired electromagnet shape in this particular situation is one that forms the shape of a xe2x80x9cCxe2x80x9d and straddles the periphery of the disc with the permanent magnets traveling through the slot. In this case, air moving vanes can be added which protrude radially past the periphery of the disc to move air within the channel of the electromagnets to provide adequate cooling. One particularly interesting approach to utilizing air cooling with this particular motor geometry is to fasten flexible plastic film vanes to the periphery of the disc. These plastic strips are initially made to protrude from the edge in a radial direction. These flexible plastic strips move modest quantities of air past the electromagnets under low RPM conditions. As the rotor velocity increases, aerodynamic drag bends these strips back. This alters their shape, reducing their drag effect on the rotor, however, enough airflow is still maintained to prevent overheating of the electromagnets.
In addition to permanent magnet motors, the large diameter planar rotor geometries employing added air moving surfaces are also suited for use in xe2x80x9cSwitch Reluctancexe2x80x9d motors. Switch reluctance motors are electric motors of brushless design where a non-magnetizable material having a high permeability is employed in the rotor in place of the permanent magnets. These materials include silicon steel, soft iron, magnetically soft ferrite, and others. Such materials become magnetic only when they are in the presence of an externally applied field. The familiar magnetic attraction between a permanent magnet and steel is an example of the principle utilized in switch reluctance motors. In such motors, only attractive forces are generated. Because of this, the switching sequence as well as the spacing of the high permeability material in the rotor is somewhat different from that which is employed in brushless electric motors having permanent magnets. Switch reluctance magnetic attraction works because while the non-magnetizable high permeability material is in the field, it becomes a temporary magnet with induced magnetic poles. Materials having high magnetic permeability lower the overall field energy of the electromagnets by completing their magnetic circuit. This field energy shows up as mechanical work. Although the risk of demagnetization is alleviated in switch reluctance motors, large voltage spikes are produced during electromagnet switching. These spikes represent large amounts of energy which can damage circuit components, create excess heat, and negatively affect motor efficiency. To alleviate this problem, high frequency chopped DC power, or even AC power can be supplied to the electromagnets where the cycle rate is considerably higher than the motor timing switching rate. In other words, during single electromagnet on times, multiple electric pulses occur.
With all of these planar rotor geometries, high surface speeds around the periphery of the rotor are easily achieved along with considerable forces in a radial direction during running. As an example, a 24xe2x80x3 diameter rotor rotating at 3,000 RPM will have a surface speed at the edge of about 200 miles per hour. With 5 pounds of permanent magnets around the periphery, in addition to a couple of pounds of other materials, about 27,000 pounds of centrifugal force would be present in a radial direction. Because of these radial forces, strong materials need to be used in the fabrication of the rotary portions of these large diameter planar rotors.
One solution is to employ composite materials such as carbon fiber. These materials are strong and lightweight. They do not easily conduct electricity like metals and therefore would not contribute to inductive losses. With some rotor permanent magnet geometries flat steel sections can be sandwiched on both sides by permanent magnets. With other permanent magnet rotor geometries, a steel band can be used for the purposes of holding the permanent magnets in place as well as magnetically connecting them in series with each other for the purposes of concentrating their magnetic flux to the desired area of the motor. Although steel is normally not a good choice due to its electrical and magnetic properties, when sandwiched between two permanent magnets or employed outside of the directly applied magnetic field of the electromagnets, it will somewhat resist the losses normally present when solid steel is subjected to rapidly changing magnetic fields. In this respect the steel used under these conditions behaves in a similar manner to the steel casing in ordinary DC electric motors. It becomes part of the magnetic circuit, however being somewhat shielded from changing magnetic fields, does not appreciably add either to inductive or hysteresis loss. For example, a hub drive system for one or more wheels in an electric vehicle can be easily employed using a steel wheel rim having permanent magnets mounted against the inside portion. The permanent magnets are placed next to each other on the inside of this rim having opposite polarity. The steel in the rim then magnetically connects these permanent magnets in series. A star shaped electromagnet is mounted in the center and bolted onto the vehicle frame. The rim portion of such a wheel is mounted to the hub portion using large structurally strong air moving spokes which provide air cooling by pumping air over the electromagnet assembly while the vehicle is in motion. This hub drive system has no gears, and therefore attaining high torque values becomes critical. In achieving this end, utilizing electromagnet windings which are capable of demagnetizing the rotor permanent magnets at stall and interlocking may be of benefit. This particular open motor design provides cooling, however such open motor designs must be made somewhat resistant to the elements.
As mentioned earlier depending on motor geometry, it may be desirable to move air by employing a turbofan as the central portion of the planar rotor. This works well when the star shaped electromagnet geometry is used in the center of the motor with the rotor permanent magnets rotating around the periphery of the centrally located stationary electromagnets. It is desirable that such a turbofan is capable of moving large volumes of air over the electromagnets while running at relatively low RPM values.
An example of such a turbofan is outlined in U.S. Pat. No. 5,075,606 by Leonard H. Lipman in which the author uses this turbofan design to move large volumes of air at low RPM values by maximizing the available cross section. The impeller portions of such a turbofan are ideal for cooling the large-diameter DC permanent magnet motor of this invention having the above described geometry.
In view of the foregoing, it is an object of this invention to provide methods and apparatus for increasing power of permanent magnet motors.
In summary, the present invention provides a method for increasing the power output of permanent magnet motors. A power supply capable of demagnetizing motor permanent magnets is power limited to a safe level sufficient to prevent demagnetization on motor start up. Speed and/or current sensing circuitry activates full power to the motor when conditions such as speed and current reach a level sufficient to prevent demagnetization.